Unger and Van Waarden 1999

Review of International Political Economy 6:4 Winter 1999: 425–467
Interest associations and economic

growth: a critique of Mancur Olson’s
Rise and Decline of Nations
Brigitte Unger and Frans van Waarden
Vienna University of Economics and Business Administration
and Utrecht University
ABSTRACT
The article contradicts the thesis of Mancur Olson on the relation between
interest associations and economic growth, presented in his book The Rise
and Decline of Nations. It is based on empirical evidence from studies on
interest associations and on sectoral corporatism. First, contrary to Olson’s
assumptions, the number of associations has not increased over time.
Instead, mergers have reduced their numbers. Nor did war break up
‘distributional coalitions’ but strengthened them and made them more
comprehensive. Second, contrary to Olson’s view, more associations do
not mean they have more political inuence, hence greater bias in public
policy. Third, interest associations are not necessarily detrimental to
economic performance and growth. By contrast, we provide evidence that
interest associations can make a positive contribution to economic perfor-
mance. This undermines the overall argument that political stability
produces more particularistic interest associations and increases institu-
tional sclerosis, thus diminishing economic performance and growth.
K E Y WO R D S
Interest associations; logic of collective action; economic growth theory;
market failures; comparative economic and political institutions; war and
institutional development.
1 IN T R O D U C T IO N
Seventeen years have passed since Mancur Lloyd Olson wrote his second
book The Rise and Decline of Nations (1982), which, according to ‘The
Economist’ (7 March 1998), could have won him a Nobel Prize in
economics. It won him at least a major award of the American Political
Science Association. The book was the rst attempt in recent history to
Review of International Political Economy

ISSN 0969-2290 print/ISSN 1466-4526 online © 1999 Taylor & Francis Ltd
REVIEW OF INTERNATIONAL POLITICAL ECONOMY
explain variations in economic growth rates by reference to institutions,

in particular interest associations or, as Olson called them, ‘distributional
coalitions’, which were supposed to have detrimental inuences on
economic change and growth. His elegant and parsimonious theory was
both intellectually challenging and aesthetically pleasing, and was
applied to a wide range of historical experiences. Now, following his
death at age 66 (on 19 February 1998), it is still an inuential work and
the theory has become almost general knowledge, as attested by the obit-
uary in The Economist: ‘Make these points to a student of economics or
politics and he or she will say, “Of course”. But the ideas were obvious
only after Mr Olson made them so’ (7 March 1998).
The work has attracted a lot of criticism (e.g. Weede, 1986; Cameron,
1988; Schubert, 1992; Maddison 1995), as is to be expected for such an
ambitious, grandiose and challenging undertaking. Much of the critique
has, however, concentrated on the growth element in the relation between
‘institutions and growth’. Thus Cameron (1988) and Maddison (1995)
come up with alternative explanations for variations in growth rates.
Cameron tries to nd these in ‘a nation’s historic and evolving position
in the world economy, the policy responses through which governments
attempt to maintain or improve that position, and, nally, the constraints
on (or, conversely, opportunities for) growth-oriented domestic economic
policy that derive from that position’ (1988: 603).
Much less critical attention has been paid to the precise nature, activ-
ities and consequences of interest associations in various countries. What
there is refers mainly to trade unions, on which whole libraries have
been written. Thus Cameron’s forty-page critique of Olson restricts its
discussion of interest associations mainly to workers’ associations and
hardly mentions cartels or trade associations, notwithstanding the fact
that Olson clearly had such associations in mind in The Rise and Decline
of Nations, as well as in his earlier book, The Logic of Collective Action
(1965). However, since the publication of his book, many new studies
of such associations have appeared. It is therefore, time for a confronta-
tion between Olson’s bold theses and the new ndings of history and
political science on the development and performance of interest asso-
ciations. That is the aim of this article.
Reference will be made to four different studies in which we partici-
pated. The rst is a cross-temporal study of the historical emergence and
organizational development of Dutch business interest associations,
including cartels (van Waarden, 1985b, 1987a, 1987b, 1989, 1992a). This
includes a le of 1,664 national associations founded between 1880 and
1960, created on the basis of contemporary association directories. The
study also includes historical case studies of the most important associ-
ations in the cotton textile, metalworking, construction, food-processing
and printing industries in this period. The second study is a cross-national
426
UNGER AND VAN WAA RD EN : A SSOCIATIONS A ND GROWTH
study on the organizational structure and activities of 352 business inter-

est associations in nine countries (Austria, Canada, West Germany, Italy,
the Netherlands, Spain, Sweden, Switzerland and the United Kingdom),
the so-called Organization of Business Interests (OBI) project (see, on the
research design, Schmitter and Streeck, 1981). Again, this project includes
both a quantitative dataset on these 352 associations and detailed country
and sector case studies. The third project is a comparative study of the
changes in the structure of business interest associations during and after
the Second World War in the context of organizing the industry to serve
the war effort in the various belligerent countries (reported in Grant
et al., 1991). The fourth and nal project was a comparative study of sec-
toral economic performance, initiated at the University of Madison,
Wisconsin (Hollingsworth et al., 1994; Traxler and Unger, 1994a, 1994b).
2 T HE LO G IC
Olson’s argument has become widely known, but to facilitate the discus-
sion we will summarize its essential features:
1 Interest associations are bad. They
reduce efciency and aggregate income in the societies in which

they operate . . . slow down a society’s capacity to adopt new tech-
nologies and to reallocate resources in response to changing con-
ditions . . . erect barriers to market entry . . . and make decisions
more slowly than the individuals and rms of which they are com-
prised and tend to have crowded agendas and bargaining tables.
(Olson, 1982: 41–69)
All this reduces economic growth. Olson subscribes here to the familiar
microeconomic picture of associations as ‘rent-seekers’ (Buchanan
et al., 1980). The supposedly malign functions of cartels also hold for
‘any combination of individuals or rms for collusive action in the
market place, whether a professional association, a labor union, a trade
association or an oligopolistic collusive group’ (Olson, 1982: 44).
2 The greater the number of such special interest associations, the worse.
Olson does not explicitly say so, but it is implicit in the argument that
stable societies allow associations to proliferate (see Section 3), which
is why their economic growth declines. As the perverting inuence of
associations is partly due to the asymmetry of interests represented –
small groups organize more easily than larger groups – this argument
also implies that an increase in numbers increases, or at least does not
decrease, the asymmetry in the system of organized interests.
3 The formation of interest associations takes time. Hence, the longer
the uninterrupted development a country has experienced, the greater
427
their number and the slower economic growth should be. ‘Stable soci-
eties with unchanged boundaries tend to accumulate more collusions
and organizations for collective action over time’ (Olson, 1982: 41, 75;
1983: 17–18). Political upheaval such as revolution, war, invasion, occu-
pation or totalitarian government could disrupt the proliferation of
such organizations. Indeed, he posits a fascinating paradox: positive
political phenomena such as stable democracy and absence of war
have their economic costs. The implication is that an authoritarian
limitation of the freedom of association may be required to safeguard
the freedom of enterprise and economic progress.
We will take up these arguments, starting with the last one and then
making our way back to arguments 2 and 1. We follow this order, like
peeling layers off an onion, because even if 3 does not hold, 1 and 2 can
still be valid, and if 2 does not hold 1 can still be valid. Section 3 discusses
the relation between time and war on the one hand and number and
comprehensiveness of interest associations on the other. In Section 4 we
address the question of whether greater numbers of interest associations
mean more overall economic and political power for such particularistic
interests. Section 5 argues that business interest associations can have
benecial economic effects. We present cases of associational self-
regulation that provide solutions to well-known problems of market
failure, thus enhancing an economy’s capacity for structural change,
international competitiveness and growth. Finally, in Section 6 we
compare long-term growth rates of countries that differ in the preva-
lence of associational self-regulation at the sectoral level.
3 T IM E, W A R A N D A S SO C I A T IO N S
3.1 Time and numbers

Does the number of interest associations keep increasing over time, as
Olson suggests? Studies on the formation of interest associations show
that their numbers grew fast at rst. Many newcomers did not survive
long because of their ‘liability of newness’ (Stinchcombe, 1965), but others
did, and over time the density of associations increased. As this
happened, the growth rate of new associations being founded slowed
down. In the meantime the disbanding rate increased, due to dissolu-
tions, but also increasingly due to mergers and transfers, which
eventually caught up with the founding rate, resulting in a more or less
stable population of associations or even a decrease in their numbers.
Table 1 lists the number of new foundings of national or nationally
relevant trade associations in the Netherlands, Canada and the USA. The
data show an increase in the birth rate of trade associations – albeit with
428
strong uctuations; we will return to that – and a certain stabilization

of the number of new foundings in the post-Second World War years
in the USA (between 210 and 270) and Canada (around fty). The Dutch
data go only to 1960 and do not immediately show a stabilization.
However, it is important to know that the relatively high numbers of
new Dutch associations in 1945–55 give a false impression. The large
majority of these foundings were not new associations, but continua-
tions of pre-war associations (dissolved by the Nazi occupation forces)
and of the compulsory trade associations created by the Nazis during
the war. That is, compulsory associations (not included in these data)
were transformed into voluntary ones. (We will return to this in the
following sections.) The difference in post-Second World War develop-
ments between Canada and the USA on the one hand and the
Netherlands on the other reects the discontinuities the latter country
experienced because of the war. The real new foundings in the period
1945–55 were much lower, so that in reality the rate tapered off here as
well.
Table 1 Foundings of nationally relevant voluntary trade associations in the

Netherlands, Canada and the USA
Period Netherlands Canada USA
1880–4 15 2 ?
1885–9 11 6 ?
1890–4 18 6 ?
1895–9 36 11 ?
1900–4 73 10 70
1905–9 99 15 76
1910–14 100 15 96
1915–19 279 49 173
1920–4 60 36 115
1925–9 76 34 108
1930–4 167 38 245
1935–9 156 37 164
1940–4 27 61 170
1945–9 249 34 229
1950–4 254 41 210
1955–9 30 33 270
1960–4 – 50 211
1965–9 – 53 218
1970–4 – 52 279
1975–9 – 39 268
Sources: for the USA: Aldrich et al. (1991); for Canada: Coleman (1988); for the Netherlands:
van Waarden (1992a).
– = no data available
429
Mortality rates of trade associations are available only for the USA
(Aldrich and Staber, 1988). They show a convergence of founding and
dissolution rates, leading to a stabilization of the number of trade asso-
ciations (cf. also Aldrich et al., 1990: 32).
Trade union foundings show a similar pattern, and here there are
more data on dissolutions. Visser and Waddington (1996) present data
on foundings, dissolutions, mergers and the average total number of
trade unions for the decades between 1892 and 1985 for Sweden, the
Netherlands and Britain. This is summarized in Table 2. The data show
a wavelike development of foundings and dissolutions, but the long-
term trend is curvilinear. As is to be expected, the number of new
foundings is high in the initial phases of unionization. After 1930 and
especially 1950 the number of new union formations decreases, most
strongly in Sweden, but also in Britain and the Netherlands, albeit with
an increase in the 1970s. In the early years, the number of dissolutions
of unions is also high, as they still suffer from the ‘liability of newness’.
As unions get established and accepted, as they learn from each other
and an associational infrastructure develops that aids newcomers, the
number of cessations decreases. However, then the number of mergers
increases, as union ofcials look for economies of scale and scope, for
securing resources and for increasing their representativeness and hence
political inuence and bargaining power.
This pattern at rst produces an increase in the total number of unions
in the three countries. However, the number peaks between 1920 and
1932. To be more precise, the highest number of unions is reached
in Britain in 1920, in Sweden in 1931, and in the Netherlands in
1932. Thereafter the number of unions decreases, most drastically
in Britain, where in 1985 the number of unions is only 24.7 per cent
of the peak number in 1920. In Sweden it is 37.4 per cent of the peak
number and in the Netherlands 68.4 per cent. Thus the size of the
reduction in numbers differs between the three countries – for various
reasons, such as Dutch pillarization – but the trend goes in the same
direction.
The number of union members increases, however, leading to a growth
in the average size of unions over the whole period: in the Netherlands
from 309 workers per union in 1892–1900 to 6,169 in 1980–5, in Britain
from 1,341 to 26,887; and in Sweden from 604 to 46,143 workers. These
averages still understate the degree of concentration that has taken place.
Of the 201 unions left in 1997 in the Netherlands, 163, or 81 per cent,
had fewer than 5,000 members. Fifteen unions, afliated to the three
main peak associations, grouped 80 per cent (1.5 million) of all orga-
nized workers (CBS, 1998). These unions were indeed the result of a
series of mergers, a trend which was stronger among workers’ unions
than among business associations, as for unions the pure number of
430
Table 2 Trade union foundings, dissolutions and mergers and number of unions (annual average per decade) for Sweden, the
Netherlands and Britain
Sweden Netherlands Britain
Period Foundings Dissolutions Mergers Number Foundings Dissolutions Mergers Number Foundings Dissolutions Mergers Number
of unions of unions of unions
1892–1900 45 2 3 55.8 112 3 4 76.1 603 322 166 1196.3

1901–10 72 4 26 96.2 190 30 50 185.5 561 363 182 1124.8
1911–20 90 4 16 143.0 281 47 116 289.1 710 228 432 1244.0
1921–30 33 1 21 197.6 132 76 65 361.9 287 311 262 1144.1
1931–40 45 2 62 209.5 63 43 40 341.8 148 155 107 1026.6
1941–50 52 3 78 195.3 175 60 75 293.2 132 132 322 832.0
1951–60 23 2 42 173.1 86 41 52 338.6 39 80 46 643.4
1961–70 9 6 72 143.6 61 37 87 304.9 111 89 70 569.4
1971–80 2 1 36 88.8 128 16 107 261.8 235 142 141 470.5
1980–5 2 0 14 79.0 20 6 88 262.8 106 90 162 402.8
Total number
for whole
period 368 25 362 1201 358 640 2794 1835 1910
Overall
average 142.2 274.1 886.5
Source: Visser and Waddington (1996).

Notes: Foundings includes new trade unions created out of breakaways from existing ones. Mergers lists the number of unions that ‘disappear’ due to a
merger: the number of unions involved in an amalgamation, and the number of unions taken over or absorbed by others. Total number lists the annual
average for the decade.
members represents a more important resource, and workers usually

have more homogeneous interests than business rms. The problem
of how to manage diversity of interests has often prevented mergers of
trade associations.
As industry initially develops and differentiates itself, so does the
pattern of economic interest associations. Newly emerging sectors of
the economy and new social categories (e.g. civil servants and profes-
sional workers, in addition to the older categories of skilled and manual
blue-collar workers) create their own trade associations and trade unions.
But there seems to be a saturation point. As the associational density
increases, the rate of foundings decreases while the merger rate increases,
leading to a decrease in the total number of associations. Aldrich et al.
(1991: 28–9) locate this saturation point for American trade associations
at 1,400. The curvilinear relation between density and founding rates
to total number of associations found by Visser and Waddington (1996)
for three European countries had already been found for American
trade unions by Hannan and Freeman (1988) for the period 1836 to 1985.
This trend is likely to continue. Both the increasing blurring of bound-
aries between economic sectors due to technological change and post-
Fordist production methods (Schmitter, 1990) and the merger trend
are likely to reduce the number of associations and the differentiation
of associational patterns even further. This is quite unlike the linear
trend of more associations with the passage of time which Olson
hypothesized.
3.2 War and disbanding of associations

What about the hypothesis that political and social upheaval caused by
war, totalitarian government or foreign occupation ‘emasculate or abolish
special-interest groups’, thus enhancing economic growth (Olson, 1982:
75; 1983: 25)? Olson sees this as an explanation for the ‘economic miracle’
in countries that suffered defeat in the Second World War, notably Japan
and Germany.
European countries that experienced severe regime changes because
of war were Germany, Austria and the Netherlands. In Germany and
Austria this happened rst with the take-over by totalitarian Nazi
regimes in 1933 and 1934, and subsequently with their defeat in 1945
and occupation by foreign powers. The Netherlands experienced regime
change with German occupation in 1940 and again with liberation
in 1945. Such regime changes should have destroyed existing business
associations. Did they?
The ndings of the OBI project indicate that they did not. Table 3
shows that Germany, Austria and the Netherlands do not have signi-
cantly younger business associations. On the contrary, Germany and
432
Table 3 Average age of business interest associations in nine countries
Number of associations Average age in

in the study years in 1985*
Relatively old associations:

West Germany 43 67
Austria 17 64
Switzerland 53 64
Medium-age associations:
United Kingdom 34 58
Sweden 33 54
Relatively new associations:
Canada 49 49
The Netherlands 69 48
Italy 15 47
Spain 18 29
Total 331 55
Source: OBI project.
Note: *Calculation of age includes any direct predecessors.
Austria have on average even somewhat older associations. Many of

their business associations trace their roots to before the Second World
War. By contrast, countries with a stable political structure such as
Sweden, Britain and Canada have relatively young business interest
associations. This might be an indication of an open pluralist system of
interest associations, which lends itself to foundings.
These quantitative ndings are supported by qualitative data from
another project coordinated by one of the authors of this article, which
shows a remarkable continuity of business associations in various coun-
tries across severe breaks in their political development before, during
and/or shortly after the Second World War (Grant et al., 1991). In only
two of the nine countries studied – Germany and the Netherlands – did
the war break up existing systems of interest associations. However, this
happened twice: at the start and at the end of the war. This ‘double
break’ meant to some extent a return to the pre-war situation. Voluntary
interest associations were re-established after the war, in part even in
their pre-war pattern of differentiation. Thus in Germany separate
employer and trade associations emerged again after the forced wartime
merger between economic and socio-political business associations (van
Waarden 1991b: 297).
433
3.3 War and crises as creators rather than

destroyers of associations
Contrary to what Olson maintains, war does not destroy interest asso-
ciations; it actually creates them. In democratic countries more voluntary
trade associations were founded during wartime. The data presented in
Table 1 show peaks in associational foundings in the periods 1915–19
(for all three countries) and 1940–4 (for Canada). Although they do not
provide quantitative data, the contributions in Grant et al. (1991) describe
an increase in foundings of trade associations during the Second World
War in Sweden, the UK, Canada and the USA. Cuff (1973) studied the
development of trade associations during the First World War and
showed how these were created to be instruments for mobilizing the
economy for the war effort. Coleman (1988) did the same for Canada,
and Grant (1987) for the UK. The data of Visser and Waddington (1996;
see Table 2 above) also show a peak in foundings of trade unions in
Sweden and the Netherlands in the war decades 1911–20 and 1941–50,
and also for Britain during the First World War and its direct aftermath.
In totalitarian Nazi Germany and the countries occupied by it,
including the Netherlands, pre-existing voluntary trade associations and
trade unions may have beeen destroyed before and during the war.
However, they were replaced by an even more extensive compulsory
corporatist system of associations for both employers and workers
(Weber, 1991; Barnouw and Nekkers, 1991). When these were in turn
disbanded after the defeat of the Nazis, the many new voluntary trade
associations and trade unions that were created – the Dutch data show
a peak in 1945–52 – were continuations of either pre-war associations or
the compulsory wartime associations.
This peak in associational foundings during wartime points to differ-
ent motives and initiators. Traditional interest group theory and political
economy, including Olson, commonly assume that initiatives for associ-
ational foundings come from the prospective membership. That is, trade
associations are assumed to be created by industry to represent their
particularistic interests, to lobby the state and to seek privileges and
‘rents’. The wartime data, however, back up the hypothesis of corporatist
theory, that associational founding has been aided by external support
from the addressees of interest representation, notably the state. Most
wartime associations were created at the instigation of the state, which
needed interlocutors from industry to cooperate with and to help orga-
nize the economy for the war effort. Many trade associations were given
public tasks, such as the distribution of scarce raw materials in line with
the priorities of the war economy, the implementation of zoning regu-
lations, or the provision of certicates of origin of products and export
licences. Once created, such associations of course became true ‘interest
434
representatives’, particularly when they lost their statutory tasks after

the war, as many did. Only in the USA were many disbanded after the
war. Elsewhere, they developed into full-edged interest associations.
This holds not only for sectoral trade associations, but also for
the general peak associations of industry. Coleman indicates that the
predecessors of the main peak associations of business in Britain (CBI),
Germany (BDI) and France (CNPF) ‘were all created between 1916 and
1920 at the behest of governments seeking assistance in dealing with the
end of the First World War. The US Chamber of Commerce was formed
at the behest of President Taft in a message to Congress on 7 December
1911’ (1990: 251).
More generally, crisis of all kinds generate associations rather than
destroying them. The importance of crises for the emergence of interest
associations has been hypothesized in the theoretical corporatist litera-
ture (Maier, 1981; Katzenstein, 1985; Lehmbruch, 1986). The present data
conrm this hypothesis. The data on the formation of trade associations
(Table 1) show strong uctuations, with peaks not only during the First
and Second World Wars, but also during times of economic crisis and
increased state intervention. Thus the American data show a peak during
the National Recovery Administration (1933–5). The Dutch data show
relative increases in the years 1900, 1906–7, 1915–19, 1932–5, 1938 and
1945–52. These were all periods of ‘crisis’ on markets, periods in which
businessmen as a result of war, depression or economic boom were
confronted with problems arising from extreme shortages on the markets
for either labour, raw materials or nal products.
The various crises were also periods of increasing state intervention
in the economy, as governments tried to alleviate the problems caused
by the crises. This induced business to organize. The First World War
brought a hitherto unknown degree of economic regulation, including
raw materials distribution, import licences, food supply and war prots.
Directly after the war, labour shortage and social unrest produced some
important labour laws. The economic crisis of the 1930s then brought
substantial market ordering legislation.
In these periods the state sponsored interest associations. Many early
Dutch employers’ associations were created at the instigation of the state,
which wanted countervailing powers to resist strikes and unions and
to prevent wage increases; or (later) associations which could negotiate
with unions and channel social conict (van Waarden 1987b, 1992a).
Furthermore, the state needed both trade unions and trade associations
to help implement public policy. It needed them for information and
assistance in the implementation and legitimation of its policies. During
the 1930s associations were created to assist the state in its market-
ordering measures. Crises and state intervention thus provided windows
of opportunity for organization.
435
3.4 War and encompassing organizations

Olson attributes the economic success of Germany and Japan (and one
might add of Austria, the Netherlands, Norway or Sweden) partly to
the existence of encompassing associations of workers and business,
which cannot so easily externalize the costs of their demands (Olson,
1982: 48–53, 76). However, Olson cannot logically explain the existence
of such associations in these countries. They appear as a deus ex machina
in his theory. If the formation of interest associations takes time, and if
small groups are privileged when it comes to organization because of
problems of collective action, how much more time is then required to
form associations of large groups or peak associations? Hence if at all,
one would expect encompassing associations in countries with a stable
political history but not in Germany, Austria or the Netherlands, where
they did actually take shape.
It turns out that war and regime change actually functioned as a cata-
lyst for the development of encompassing associations. War may not
have produced a break-up of distributional coalitions. It did, however,
spawn modications of the systems of interest associations. Wartime led
frequently to a ‘redesign’ of associational systems. Interest associations
became larger and more comprehensive. Thus the German, Austrian and
Dutch trade union movements were reorganized twice. When the Nazis
came to power they abolished the unions and established a compulsory
and authoritarian unitary organization, the Arbeitsfront. At the end of the
war this organization was dissolved by the Allied forces. A new pattern
of union organization was established, which could be much more com-
prehensive than what existed before the Nazi period, precisely because
the Nazis had destroyed the former differentiated structures (Weber,
1991; Barnouw and Nekkers, 1991). Germany’s present comprehensive
organization of small artisan rms in the Zentralverband des deutschen
Handwerks and its afliated chambers and guilds, though rooted in
age-old guild traditions, was the creation of the Nazis in 1933 and sur-
vived post-war liberalization (Streeck, 1989). Independent pre-war trade
associations, which had been forcibly integrated into the system of
Nazi Wirtschaftsgruppen, became sectoral chapters in industry-wide
associations after the war. The wartime experience also increased the
centralization of interest associations in France and the Netherlands.
Earlier on, the First World War had produced a similar effect. In neutral
Sweden, the surrounding war and threat of state intervention forced the
peak trade union LO to centralize its internal decision-making processes,
an old demand of the already centralized peak employers’ association
SAF, which wanted an equal negotiating partner (de Geer, 1991; Flam,
1990: 26–7).
This shows up in data on the number of peak associations of workers
and the share in total union membership of the largest peak association
436
Table 4 Concentration of the union movement: number of peak union

organizations and share in total union membership of the largest peak before
and after the Second World War
Country Number of peak Union Total union member-

unions ship of largest peak
union (%)
1933 1939 1945 1985 1933 1939 1945 1985
Axis/occupied countries:
Austria 3 – 1 1 BFG (1894†) 77 –
OGB (1945) – 100 100
Germany* 8 – 4 5 ADGB (1875) 67 (1931) – –
DGB (1950) – – 92
Netherlands 8 8 6 7 NVV (1905) 40 – 32
FNV (1975) 58
France 4 3 3 6 CGT (1895) 58 89 93 32
Norway 1 1 1 3 LO (1899) 100 100 100 67
Denmark 1 1 1 5 LO (1898) 74 94 95 70
Allied countries:
UK 1 1 1 1 TUC (1868) 75 77 85 89
Neutral countries:
Sweden 4 5 4 4 LO (1898) 84 86 82 60
Switzerland 5 5 5 4 SGB (1880) 60 58 60 51
Source: Visser (1989).
Notes: * Data on Germany in column 1945 are for 1950, as the union organization was
only re-established in that year.
†
Founding dates are given in parentheses.
(Table 4; data from Visser, 1989). The data show a higher concentration
of the union movement in comprehensive peak associations shortly after
the Second World War than there was before for the Axis powers, Austria
and Germany, and two of the countries they occupied, France and the
Netherlands. The number of peak associations was reduced from three
to one (Austria), eight to four (Germany), four to three (France) and eight
to six (the Netherlands). The share in total union membership of the
largest peak union increased from 77 to 100 per cent (Austria), 67 to 92
per cent (Germany), and 89 to 93 per cent (France). Only in the
Netherlands did it decrease, from 40 to 32 per cent. As for the other Axis
power, Italy, Visser does not provide data, but writes: ‘After the defeat
of fascism, the Italian trade union movement was reconstituted in a uni-
tarian fashion (Pact of Rome of 1944) into the Confederazione Generale
Italiana del Lavoro (CGIL)’ (1989: 118). There was also some effect in the
countries that were neutral or that did not ght the war on their own
soil. In Sweden the number of peak unions decreased from ve to four,
437
and in Britain – where there had always been just one peak union, albeit
it a weak one – the TUC’s share of total union membership increased
from 77 to 85 per cent. Only in Norway, Denmark (both with different
German occupation regimes) and Switzerland did the war have no
discernible effect on union concentration. Table 4 also shows that later
in the post-war years union concentration decreased again (more peak
unions, lower share for the largest), as the associational pattern of unions
split along political or religious lines (Italy, France) or along occupational
ones, with new peak associations for white-collar workers, professionals
and civil servants (Netherlands, Germany, Scandinavia).
Furthermore, several post-war comprehensive ‘distributional coali-
tions’ between the peak organizations of labour and of capital had their
roots in the wartime experience. In all countries except France and
Canada, the position of trade unions was strengthened. They were
recognized on a permanent basis by the state and by the employers’ asso-
ciations as partners in consultations and negotiations. In Britain and other
democratic countries this was due to their participation in wartime advi-
sory and implementation committees. In the occupied countries it was
mainly a result of close cooperation with the underground and resistance
movements (see also Katzenstein, 1985: 147; Maier, 1981: 48). Several
macro-level forms of concertation emerged from this. The Austrian
Sozialpartnerschaft and the political accommodation between the erce
pre-war adversaries, the Christian Democrats and the Social Democrats,
were born out of the experience of fascism, war and occupation. The
Dutch bipartite Foundation of Labour, and indirectly its successor, the
tripartite Social-Economic Council, were formed during informal con-
sultations between representatives of labour and capital in the under-
ground during the war. The presence of a common enemy brought the
antagonistic parties together (e.g. Windmuller, 1970).
Sectoral trade associations were also strengthened rather than weak-
ened by the wartime experience. Many post-war sectoral business
associations in Germany emerged out of Reichsgruppen of the Organi-
zation der Gewerblichen Wirtschaft, the comprehensive and hierarchic
system of compulsory organizations which the Nazis created and in
which many pre-war business associations merged (Weber, 1991). Similar
developments took place in France, Belgium and the Netherlands, where
the German occupants had introduced a comprehensive form of state
corporatism comparable to that in their own country (Rossiter, 1991;
Luyten, 1991; Barnouw and Nekkers, 1991).
To sum up, in countries which experienced neither fascism, totalitar-
ianism nor intensive occupation – Britain, Canada, the USA, Sweden,
Denmark – the war did not leave any lasting imprint on the structure
of interest associations. As Olson assumed, it made a greater impact on
associational systems in countries like Germany, Austria, the Netherlands
438
and France. However, contrary to Olson’s assumptions, the war did not
weaken but strengthened interest associations, and actually helped create
encompassing organizations. It did so because it intensied the state’s
need for assistance from private associations, motivating state support
for interest associations in various forms.
4 D O ‘M O R E’ A S SO C IA T IO NS ME A N MO R E
PA R T IC U L A R ISM ?
4.1 Numbers and political inuence

As shown above, there are many data that cast doubt on Olson’s theses
that over time more distributional coalitions are formed and that war
and crisis interrupt this process. However, his argument that the greater
the number of distributional coalitions, the more harmful they are, might
still be valid.
Interest associations exert their inuence on the economy in large part
indirectly, through their inuence on public policy. Steel interests lobby
for tariffs to protect them from foreign competition; the pharmaceutical
industry lobbies for longer-lasting patents to limit market entry by other
producers. Arguing that the more interest associations, the worse their
economic effects is like saying: the greater the numbers, the greater
their political inuence, or the more public policy is dictated by partic-
ularistic interests.
This militates against everything political science has discovered about
the inuence of interest groups. In fact, beyond a certain threshold there
is probably an inverse relationship between numbers and inuence of
interest associations. State agencies that nd only one dominant interest
organization in their environment are more easily inuenced by it than
agencies that can play off various opposing interests against each other.
Studies on the different so-called ‘independent regulatory agencies’ in
the USA conrm this. The ‘economic’ independent regulatory agencies
in the USA which were founded from the 1890s on, such as the Interstate
Commerce Commission, the Federal Communications Commission, the
Federal Aeronautics Board and the state agencies that controlled public
utilities, usually regulated only one industry and found one major inter-
est organization – a rm with a natural monopoly or a dominant trade
association – in their environment. Studies of these agencies have shown
that they were rather prone to ‘capture’ by their clientele (general stud-
ies are: Bernstein, 1955; McConnell, 1966; Mitnick, 1980; on the ICC, e.g.
Hoogenboom and Hoogenboom, 1976). The ‘social’ regulatory agencies
founded from the 1960s on, such as the Environmental Protection Agency
(EPA) or the Occupational Safety and Health Administration (OSHA),
were not specialized in any one industry, and hence found themselves
439
confronted with interests from many economic sectors. By balancing

these interests and changing alliances, such regulatory agencies were
able to keep a much greater autonomy than in a clientelistic environ-
ment. They did their name of ‘independent regulatory agencies’ more
honour than the older economic ones (albeit that the term ‘independent’
in the name referred to a different independence, an independence
from the president and its spoils policy (Shapiro, 1997)). It is indicative
of this autonomy that the inuence of the opposing interests of busi-
ness on the one hand and the environmentalists and labour on the other
varied over time with changes in the political climate. This variable
inuence is reected in the title David Vogel gave to his study of the polit-
ical inuence of business on the social regulatory agencies, Fluctuating
Fortunes (1989).
As the number of interest associations increases, clientelistic policy
environments become transformed into pluralistic ones. In these, indi-
vidual interest associations will have much more difculty in asserting
their interests. In sum, contrary to what Olson maintains, it could be
that the smaller the number of associations, the greater their political
inuence; hence the more likely that public policy will be biased to
those special interests. Where these few associations have become fairly
comprehensive, the inuence they exert will of course favour more
general interests, leading to less ‘rents’ for only special interests.
4.2 Asymmetry in the political arena?

The traditional pluralist argument is that it is easier for the state to main-
tain its neutrality and autonomy if it is confronted with a plurality of
competing interests. This argument assumes, however, a minimal degree
of symmetry and balance between different interests, which is to say the
least not guaranteed. Mancur Olson criticized this assumption. He
presented compelling reasons why interest groups should be unequal in
their capacity to organize as powerful lobbies. Small homogeneous
groups are ‘privileged’ in this respect, because they can overcome free-
rider problems more easily than large groups. Thus businessmen, who
usually form small groups, should nd it easier to organize than large
heterogeneous groups such as those of workers, consumers or sufferers
from environmental pollution. Indeed, evidence abounds that density
ratios of business interest associations are much higher than those of
trade unions, consumer organizations or environmental groups. Even
among business interest associations, there is an inverse relation between
size of the recruitment pool and the density ratio (van Waarden, 1991a).
Nevertheless, some arguments can qualify Olson’s assumption of an
asymmetry between interests in the political arena, which in part can be
derived from his own logic.
440
As by Olson’s assumption, more associations emerge over time, the

pluralistic ideal would be ever more closely approached. The more
interest groups are able to form organizations, the more interest associ-
ations would nd other organizations in their environment, with
different, or even opposing, interests. State agencies would be better able
to maintain autonomy from these different interests, reducing the inu-
ence of interest groups on public policy.
Even if one assumes that large groups in particular, such as consumers
or environmentalists, are never able to organize effectively, this does not
imply that business interests would not nd powerful opponents in the
political arena. The ercest opponents of specic business interests are
usually not consumer interests but rival business interests. Thus the steel
industry may want tariffs on steel imports, but the automobile industry
is very much against them. Firms dependent on domestic demand have
different interests from rms which export a large percentage of their
output.
Business is more heterogeneous than any other group. This structural
heterogeneity is reected in the differentiation of industry into many
sectors and branches. This differentiation is an advantage for initial orga-
nization, as Olson has argued (Olson, 1965: 145; see also Moe, 1980: 197).
But it produces a highly fragmented system of often competing and
opposing interest associations.
As against Olson’s argument that particularistic business associations
would have more political power because they have a ‘privileged’ capac-
ity for organization, one could argue that density ratios and other orga-
nizational resources are not the only factors determining political clout.
Political inuence seems, especially in pluralistic policy worlds, more
dependent on other factors such as the business cycle, unemployment,
political culture, the fashionable political issues of the day and, more
generally, the importance of a social group for the realization of specic
state goals. Furthermore, access and inuence depend also on lters built
into the legal, political and administrative state institutions. These differ
between countries. Policy studies have shown that public administrators
in Britain, Germany, Sweden and the Netherlands tend to make them-
selves accessible to interest groups whereas French and American civil
servants tend to be much more suspicious of special interests and try to
keep them at bay (Kelman, 1981; Wilson, 1986; Vogel, 1986).
To sum up: an asymmetry in the capacity to organize does not neces-
sarily imply an asymmetry in actual political inuence. It also depends
on other factors. This is not meant to imply that symmetry will be the
result. But as Olson assumes, a pluralistic ideal could be approached as
more and more special (business) interest associations are created.
One could add another critical argument: as more interest associa-
tions enter the political arena, they are likely to join in the scramble for
441
access, recognition, inuence, concessions, etc., thus uprooting possibly

established relationships between interest associations and state agen-
cies. Why should what Olson holds to be the case for the economic
market – low or no market-entry limitations foster dynamism, exibility,
better resource allocation, etc. – not hold for the political market? Do
continuous new entrants not make it a more competitive – that is, plural-
istic and symmetric – market? The greater the political stability over
time, the more interest groups should have been able to organize, and
hence the more competitive the political market should be.
5 P OS IT IV E EC ON O MI C EF F EC T S O F IN T ER E ST
A SS OC I A T IO N S: C OR R EC T IN G F O R M A R K E T
A N D ST A T E F A ILU R ES
Even though Olson may be mistaken in maintaining that more associa-
tions, created over a longer politically undisturbed time period, make
for even more particularism in public policy, less efciency and lower
economic growth, his basic thesis that interest associations are detri-
mental to economic growth might still hold. However, this argument is
also questionable. At the very least it is one-sided.
It is a familiar criticism that special interest associations allow for the
extraction of undeserved rents and the externalization of costs, through
higher prices or poorer products for consumers, and that they slow down
economic decision making, reduce the pressures of competition, retard
technological development, and more generally produce rigidities and
inexibilities, whether through their own action (by cartels etc.) or indi-
rectly (by lobbying, and thus biasing or perverting public policy).
The negative view of interest associations stems partly from the
assumption that they are only or mainly concerned with redistribution
rather than with production. Olson sees associations as being ‘over-
whelmingly oriented to struggles over the distribution of income and
wealth rather than to the production of additional output’ (Olson, 1982:
44). The benets they acquire for their members are necessarily costs to
others. But would it not be possible for associations to represent
members’ interests without externalizing costs to others? Does the
political world necessarily have to be zero-sum?
This view on interest associations may be true for America. Anti-trust
legislation (Lindberg et al., 1991), adversarialism (Marcus, 1984; Chandler,
1980) and mutual distrust between business and the state (Vogel, 1978)
restrict American economic interest associations to the role of lobbying,
pressuring state agencies on behalf of narrow interests. This intensies
the critical attitude in American popular culture towards ‘special’ inter-
ests, which in turn reinforces adversarial relations and the orientation
towards redistribution. However, many European interest associations
442
do assist production and efcient allocation. They do so in particular

through the correction of market failures.
Economic theory has accepted that intervention in the market through
regulation may be necessary to correct market failures. Well-known
causes of market failures and inefciencies are information and power
asymmetries, externalities, transaction costs, ruinous competition, conict
escalation, underproduction of collective goods such as vocational train-
ing, and temporal market failures resulting in long-term inefciencies.
Market regulation is often provided by the state, because it suffers least
from free-rider problems. Thanks to its monopoly over the legitimate exer-
cise of force, and hence its monopoly over the enactment of binding leg-
islation, it can coerce prospective free riders. However, there is not only
‘market failure’ but also ‘state failure’, as many writers on public admin-
istration and regulation have documented. There may be problems with,
for example, implementation and enforcement. Policy makers cannot be
sure that ‘street-level bureaucrats’ will implement the rules according to
their intention, the familiar principal–agent problem. State agencies may
lack the required resources to guarantee ‘full enforcement’. Furthermore,
state regulation may lack legitimacy among the subjects of regulation,
who hence may oppose it and appeal against the rules and decisions in
court, leading to lengthy and costly court battles. Not only may this
lengthen the period of implementation considerably (that is, delay deci-
sion making); it may also produce extensive case law, thus increasing the
complexity of and controversy over regulation, giving subjects a sense
of ‘regulatory unreasonableness’ and ‘overregulation’, which further
increases opposition to regulation (Bardach and Kagan, 1982; Vogel, 1986).
Market failure and state failure both being possible, several European
nations have gained good experiences with market regulation through
associations. Self-regulation by industry has at least the advantage that
implementation is closer to the subjects to be regulated, so account can
be taken of the specic conditions and the constraints of the industry.
Furthermore, self-regulation tends to increase the legitimacy of regula-
tions among the subjects of regulation, thus reducing controversy and
opposition and the attendant costs.
Because of free-rider problems, industries have at times needed to call
for state backing for self-regulatory arrangements. Where this has
happened, either passively in the form of ofcial tolerance, or actively,
in the form of statutory sanctioning, it has produced mixed
public–private forms of regulation, which have also been called meso-
or sectoral corporatism.
In the following sections we discuss several typical market failures
and present examples of associations which have provided regulations
or collective goods to counter such failures, thus contributing to economic
efciency and growth.
443
5.1 Information asymmetries, opportunism, distrust:

product quality control in the agro-food industries
and future delivery services
Fierce competition in unregulated markets may entice, if not force,
competitors or transaction partners to engage in opportunistic behav-
iour. This may create or intensify mutual distrust and lead to high
transaction costs.
Distrust and opportunistic behaviour may be furthered by information
asymmetries. These may tempt entrepreneurs to engage in the adulteration
of product quality (or, for that matter, to hide the dangers of specic work-
ing conditions). This temptation is particularly great in generic products,
for which it is difcult to built up individual brand names which would
facilitate product accountability of rms. Examples are meat, milk, cheese,
vegetables or certain medicines. Meat does not show whether hormones
have gone into its production, nor butter whether margarine has been
mixed with it. It is only after the transaction has been made that one may
become aware of the consequences. Under such conditions, erce market
competition may induce manufacturers or traders to adulterate; for
certain products, such as food, toys, medicine or buildings, this may
endanger the health and safety of consumers. In such cases most gov-
ernments have eventually set product quality standards.
However, trade associations have also taken on tasks in product quality
regulation and standardization, sometimes alone, sometimes in cooper-
ation with the state, the latter particularly where public health issues
were involved. In several European countries, state regulations are codi-
cations of earlier private regulations. Examples abound. A good one
is Dutch dairy quality regulation. Quality became severely threatened
around 1900. Owing to the presence of a well-developed margarine
industry, Dutch butter manufacturers frequently mixed margarine with
the butter – so much that the mixture became known internationally as
‘Dutch butter’. Furthermore, cheese was made from skimmed milk
without consumers being informed. The cheese appeared normal while
it was fresh, but with ageing it collapsed. In 1903 there was an infamous
court case in England, involving a Dutch ‘civil engineering work’ of only
1.6 per cent fat and 57 per cent water. Such fraud was detrimental not
only to the consumer but also to the reputation of the Dutch dairy
industry. Exports dropped to an all-time low. Individually, rms could
do nothing to stop it, owing to free-rider problems. They therefore formed
associations, which set quality standards and tried to control the trade
through monopolizing the butter auctions and boycotting unreliable
merchants. However, in time they could not function without public
coercion, which came when the state required formal butter or cheese
marks for export licences. The implementation of these schemes is,
444
however, still in the hands of private interest associations. Similar devel-

opments took place in many other sectors of the Dutch economy.
Associations are institutions that help to create trust and thus facilitate
transactions. This applies in particular to transactions of goods to be
delivered in the future. Here the trust problem is especially great. How
can the buyer of an insurance plan be certain that the insurance company
will still exist twenty-ve years from now when benets are to be
paid out? Or that the travel agency won’t go bankrupt before the start
of the trip purchased? Such trust – and hence the transactions – require
a third party guaranteeing observation of the contract. Usually the
state supplies such a service. It registers, sets standards for, and controls
banks and insurance companies. More generally, it backs up contracts
by civil law.
However, trade associations have also assumed responsibilities in this
area. Thus Dutch and Austrian associations of travel agencies, car repair
shops and construction rms have created funds that guarantee
consumers delivery of the goods by the sector if a specic supplier goes
bankrupt before delivery. By advertising such guarantee funds, associ-
ations encourage consumers to demand guarantees for their future
delivery contracts and hence force rms to associate with the funds and
to satisfy their conditions. Often such collective guarantee schemes have
been combined with private certication of rms to be admitted to the
guarantee fund. Thus they set standards for nancial solvency, profes-
sional training, or the quality of goods and services provided.
Furthermore, such standards are backed up by complaints procedures
and controls on the quality of work provided. The Dutch car repair and
body shop associations investigate customers’ complaints about the
necessity or quality of repairs provided by their members. In this way
they raise the quality and reliability of rms operating in these markets.
This in turn contributes to trust between transaction partners and to
increased transactions.
5.2 Reduction of transaction costs of commercial

and social conict
By increasing trust between economic actors, associations help to reduce
the – potentially high – costs of resistance to legislation and of commer-
cial, social and legal conict. Conict and all-out competition in markets
certainly have advantages which are too well known to deserve
mentioning (keeping economic actors alert, stimulating innovation, cost
cutting, efciency, etc.). However, competition and conict, whether
between competitors, between customers and suppliers, employers and
employees, or between rms and the state also have their own costs
which may become substantial: duplicate investments, boycotts, strikes
445
and lengthy court battles. These transaction costs may be considered

socially inefcient.
Associations may help to reduce these costs. They may do so by gener-
ally increasing trust. Furthermore, specic associational activities may
help to prevent conict; where this cannot be avoided, they provide
alternative and cheaper procedures for conict resolution, such as
(continued) negotiations and arbitration. Self-regulation, mentioned
above, usually helps to build a certain degree of cooperation among the
subjects of regulation, thus reducing opposition to regulations, including
administrative and civil litigation. This saves on the lawyer costs.
Collective bargaining between associations of customers and suppliers
and between those of workers and employers may channel and mitigate
commercial and social conict and prevent boycotts and strikes. Informal
mediation procedures, arbitration clauses in associational regulations or
contracts among different associations, and specially created arbitration
boards keep commercial and employment conicts out of the courts.
Such procedures are often more efcient in settling legal conicts, both
because they involve less costs in lawyers’ fees and damages and because
they may help reduce conict between the partners involved, whereas
litigation often tends to escalate already existing conict.
Dutch trade associations and trade unions have many such provisions
and arbitration bodies that help both to prevent and to resolve social
or commercial conict. As already stated, travel agencies and construc-
tion rms have formed guarantee funds to indemnify customers
against the sudden bankruptcy of their suppliers. That implies that
customers do not have to resort to tort litigation for compensation. Other
trade associations, in such diverse sectors as construction, the dock-
yards, mineral oil markets, travel agencies and marriage counselling
bureaux, have formed arbitration boards. These boards have a bi- or
tripartite composition, with representatives of suppliers and customers
/consumers or employers and employees. Commercial disputes have to
be submitted to these boards. They try to bring both parties to a volun-
tary solution, try to satisfy both parties, and if necessary take a once
and for all decision, preventing a lengthy and costly voyage through
the court system. As a result, settlement of trade disputes is much
more efcient, as is illustrated by a case study in Kagan (1990): ‘Claims
agents who deal with cargo damage disputes arising from trans-Atlantic
shipments say that lawyers’ bills are far higher if a legal dispute is
processed in New York rather than in Rotterdam, even though the
relevant substantive law in the two countries is essentially identical.’
Kagan concludes that ‘American litigation, with its wasteful, lawyer-
dominated pretrial discovery and its cumbersome jury trials, is far less
efcient than Continental European or British methods’ (Kagan and
Axelrad, 1996: 10).
446
Other facilities that reduce the costs of (tort) litigation include unem-
ployment benets and job protection provisions, agreed to in negotiations
between employers’ associations and trade unions. An illustration is the
incidence of asbestos-based tort cases. Although the frequency of
asbestos-related diseases among Dutch workers was ve to ten times as
high as in the USA in the 1970s and 1980s (Kagan and Axelrad, 1996:
5), Dutchmen rarely went to court. By 1991 fewer than ten cases had
been led (Vinke and Wilthagen, 1992), although Dutch law authorizes
tort claims against employers. By contrast, an estimated 200,000 asbestos
tort cases had been led in the United States. The explanation is that
victims had other roads open to them in the Netherlands. For about a
century, compensation for damage caused by work accidents and
diseases has predominantly been based on social security. Financial
consequences of the risks of labour were collectivized. The same holds
for many other European countries.
As a result of such conict-mitigating associational provisions, litiga-
tion costs are much lower in the Netherlands than in many other
countries. The difference becomes particularly striking in comparison
with the litigious USA. Table 5 shows that the USA has a lawyer density
ten times that of the Netherlands: 312 lawyers per 100,000 inhabitants
as against 35. Expenditure on the professional services of the 780,000
lawyers in the USA has been estimated to be more than $100 billion a
year, or 2.4 per cent of GDP (Sander, 1992, quoted in Kagan and Axelrad,
1996; see also Lipset, 1996: 50). This made the US legal industry larger
than the US steel or automobile industry measured in terms of value
added. In the larger West European nations, the costs of going to court
amounted to 0.5 to 0.6 per cent of GDP. In the Netherlands they were
negligible.
Associations may help to reduce not only legal conict but also under-
lying social conict such as labour unrest. Potential social conict is thus
Table 5 Density of lawyers and tort cases by country
Country Lawyers Lawyers per Tort costs as %

100,000 population of GNP, 1987
United States 780,000 312.0 2.4

West Germany 115,900 190.1 0.5
England and Wales 68,067 134.0 0.5
Italy 46,401 81.2 0.5
France 27,700 49.1 0.6
Netherlands 5,124 35.2 –
Sources: ‘The rule of lawyers’, The Economist Survey, 18 July 1992, pp. 3–4, and ‘Order in
the tort’, The Economist Survey, 18 July 1992, pp. 10–13, quoted in Lipset (1996: 50).
Note: Data on lawyers for ‘last available years’ (in 1992).
447
channelled into institutionalized consultation and collective bargaining,

resulting in substantially fewer strikes; this increases productivity, one
more way in which associations may be involved in ‘production’ and
not just ‘redistribution’. This is demonstrated in Table 6, which shows
that countries which score high on associational collective bargaining
score low on strike frequency.
In Table 6, a number of OECD countries are grouped in four cate-
gories (taken from Visser, 1998 and Traxler, 1996). Category I is that of
‘Northern corporatism’, where wage bargaining is done at a central level,
by national sector-unspecic peak associations. Category II is called
‘Central social partnership’. In these countries wage bargaining is
primarily done at the sectoral level between sectoral employers’ associ-
ations and trade unions, which, however, each coordinate their policies
within less centralized peak associations. In both these categories the
associations of both sides are relatively strong in the sense that they
have a medium to high density ratio and that they are well-resourced
organizations recognized by and given access to relevant interlocutors.
Relations between social partners are usually integrative rather than
adversarial. That is not the case with category III called ‘Latin confronta-
tion’. Here the associations are relatively weak (low density ratio, limited
organizational resources, little recognition by or access to state agencies),
and relations tend to be confrontational. They try to bargain but success
is limited, leading to frequent state intervention in wage setting. The
fourth and last category is called ‘Anglo-Saxon pluralism’, with mostly
rm-level bargaining by usually relatively weak trade unions.
Table 6 shows that over a long period the greater the role of relatively
independent sectoral associations in wage bargaining, the lower the strike
frequency. Lowest strike rates are found in the countries with strong
sectoral associations that do the wage setting. Where the peak associa-
tions are strongest and most active (to the detriment of sectoral
associations), strike rates are already higher. Apparently these central
associations lack the legitimacy and authority among workers to prevent
strikes. The highest strike rates are found in those countries where
associations play a minor role in wage setting, in the southern European
countries where in the absence of strong associations the state plays an
important role, and in the liberal pluralist Anglo-Saxon countries, where
most bargaining takes place at the rm level.
5.3 Short-termism and the problem of providing collective goods

A major deciency of the market is that it often fosters business policies
which may be rational in the short run but may result in long-term
inefciencies. While long-term investments are required to maintain long-
term international competitiveness, rms may not make such investments
448
Table 6 Associational collective bargaining and strike frequency
Category and country % workers covered by Working days lost to strikes

collective wage agreements per 1,000 workers
(annual average 1970–95)
I Northern corporatism: centralized bargaining:

Sweden 83 105
Finland 95 452
Norway 75 68
Denmark 74 179
Average 201
II Central social partnership: sectoral bargaining with some central coordination:
Austria 98 6
Netherlands 71 29
Germany 82 35
Switzerland 53 1
Average 18
III Latin confrontation: weak associations, contestational social relations, large role for
the state in wage setting:
France 92 173
Italy n.a. 1,095
Spain 71 661
Average 643
IV Anglo-Saxon pluralism: uncoordinated bargaining at company/rm level:
UK 47 354
Ireland n.a. 469
USA 18 196*
Average 340
Sources: Categorization of countries is based on Visser (1998) and Traxler (1996). Data on
coverage rate from Traxler (1996) (adjusted coverage rate, i.e. categories of workers, for
example, public employees, who have no right to collective bargaining are excluded). Data
on strike frequency from Visser (1998) (based on ILO data). US data from Visser and
Hemerijck (1997).
Note: * Data strike frequency is average for 1960–95. n.a. = not available.
because they might reduce their short-term prots. Firms have to take
account of what their competitors do in the short run. Thus long-term
investment planning may be guided by short-term competitive positions
on markets. This danger increases the greater and more structural the
uncertainty about the success of the investments (in the end marketable
goods and future sales prospects), the higher the sunk costs of long-term
investment, the greater the importance of stock markets and investors/
shareholders interested in short-term prot, and the greater the dif-
culties with the private appropriation of the fruits of investments.
449
A special case arises where it is difcult to exclude others from sharing

in the benets of a private investment. This basically means private
investment in public goods. Cases are generic research and development
and training. The threat that trained workers may be poached and that
R&D ndings may be copied by competitors tends to discourage
managers from making investments which may be necessary for the
long-term competitiveness of the rm, the sector and the country. This
danger is especially great where the assets are not very rm-specic,
where labour is traditionally relatively mobile, where knowledge is more
codied than tacit, and where it is difcult to patent or keep knowledge
otherwise secret and private.
The usual answer to this problem is government nancing or even
organization of basic/generic R&D and training in public vocational
schools, universities and public research institutes. This has two disad-
vantages. First, training in such settings is often rather far removed from
actual practice in industry. Second, all taxpayers share in the costs of col-
lective goods whose benets go, at least directly, only to a small segment
of the population, specic industries and their (prospective) workers.
Both disadvantages can be countered by having such industries nance
and organize sector-specic training and general R&D themselves.
That is exactly what has been done in several countries, with trade
associations playing an important role. Because they are instrumental in
providing collective goods, the state has endowed them with sometimes
statutory powers: compulsory membership for rms in the sector, the
right to levy fees for sector-specic goods, a monopoly on the provision
of such goods. Of the 352 business associations in nine European
countries studied in the Organization of Business Interests research
project, 35 per cent had programmes for collective research and devel-
opment, and 47 per cent engaged in vocational training programmes for
workers and/or for the entrepreneurs themselves, thus raising the quali-
cation levels of workers and entrepreneurs.
Perhaps the best-developed systems are those of the German, Austrian
and Swiss regional Chambers of Artisans, of Industry and Trade, and
of Agriculture, and the sectoral guilds that are afliated to them. They
are backed by statutory powers, as in the German Handwerksrecht,
providing them with compulsory membership to combat the problem
of free riders, the right to levy fees and set training standards, the
monopoly on conducting exams, etc. Furthermore, training requirements
for the establishment of rms or shops force prospective entrepreneurs
to take part in these courses. The voluntary trade associations that exist
in addition to these compulsory ones are also closely involved in either
their governance or their activities. Thus, for example, they may organ-
ize training courses or set exams. They provide the so-called ‘dual
system’, combining theoretical and practical training, for which the
450
Germans are so famous. This system of trade associations is responsible

for the high skill level of workers, one of the assets of German industry
and trade (Streeck, 1989; Crouch, 1995).
The Dutch equivalent of the German, Austrian and Swiss Chambers
are the statutory trade associations (STAs), compulsory associations with
the power to enact binding regulations, enforceable under public admin-
istrative law, and the power to ‘tax’ their sector to nance sectoral col-
lective goods. They can be created at the request of a sector and are found
in about a quarter of industry, mostly the agro-food sector. The STAs
nance and organize collective goods, such as product quality control
(mentioned above), vocational training, research and development, and
collective advertising. These forms of associational self-regulation
and service provision have contributed to the strong competitive posi-
tion of the Dutch agro-food sectors in world markets, as is testied by
their export performance. Most sectors that scored high on the list of the
Dutch sectors with a high share in world trade are sectors organized in
statutory trade associations. Such sectors are the producers of cut ow-
ers (share in total world trade of 64 per cent), bulbs and plants (56 per
cent), eggs (61 per cent), pigs (57 per cent), condensed milk (53 per cent),
cacao powder (49 per cent), tomatoes (43 per cent), raw potatoes (36 per
cent), pork (32 per cent), cacao butter (32 per cent), beer (30 per cent),
tobacco products (28 per cent), salted pork (25 per cent), margarine
(24 per cent), milk powder (23 per cent), chicken meat, butter and cheese
(each 22 per cent), starch (21 per cent), and seeds (20 per cent) (data from
UN Trade Statistics, quoted in Jacobs et al., 1990: 29).
5.4 Ruinous competition: cartels in the construction industry

The most ideal-typical cases of ‘distributional coalitions’ are cartels. It
seems almost a commonplace that these are ‘bad’ for economic efciency,
dynamism, innovation and growth. But are they really always so detri-
mental and inefcient?
A well-known recipe for market failure is ruinous competition, as can
be illustrated by the construction industry. Here market entry and exit
are relatively easy, due to the low investments required. Although
competition may force less efcient producers out of the market, they
may come back the next day under a different name. Bankruptcy does
not cost much. Easy entry and exit make for a high turnover of rms
and for large numbers of small rms, reducing the transparency of the
market and, concomitantly, the possibility on learning effects on credi-
tors. Permanent overcapacity leads to erce and destructive competition,
resulting in prices below average costs and to pressure on workers to
accept low (black) wages and dangerous working conditions. The high
turnover is illustrated by the relatively unregulated British construction
451
industry, where up to 40 per cent of rms disappear every year (Barclays

Bank, 1990). The problem of ruinous competition is, however, not limited
to the construction industry, but has historically also been found in
sectors like printing and retailing.
Fierce ruinous competition may also be aggravated by imbalances of
power between customer and supplier or by tendering procedures. These
pit one customer or principal against several suppliers, allowing the prin-
cipal to play off prospective contractors against each other and forcing
prices down. To offset such dependencies and to ght permanent over-
capacity, some national construction industries have organized trade
associations and cartels. The Dutch industry, for example, had organized
a prototypical ‘distributional coalition’ or cartel. This was a highly visible
and controversial cartel, which has been persecuted by the European
Commission and condemned by the European Court of Justice.
However, it is a rather unambitious cartel. It does not x general prices
or conditions of supply. Nor is it a quota division agreement. It merely
provides a system of individual price xing. Members of (regional) cartel
associations are obliged to report any intention to offer a price to a prin-
cipal. When two or more contractors intend to submit tenders for the
same work, the cartel has to organize a pre-tender meeting at which all
interested members must be present. The meeting can decide to ‘deter-
mine the rightful claimant’. All participants then have to submit their
price in writing to the chairperson. The latter determines who has the
lowest price and thus the right to the work. Members may also decide
to make all prices public at the meeting, to give the participants an
insight into price development. All participants are bound by their prices.
These are the ones they have to submit to the principal, increased with
any surcharges agreed to at the meeting. They are not allowed to bargain
subsequently with the principal for a specied term. Thus the cartel
should prevent principals from playing off suppliers against each other
and bargaining a second round, confronting contractors with the prices
of their competitors.
This construction cartel was unique in Europe (Czarnovski et al., 1976;
OECD, 1976, 1978). The Dutch allowed what is forbidden elsewhere.
Various other countries had less far-reaching regulations. In England,
prices were compared through ‘builders’ conferences’ after tenders had
been submitted in order to give contractors an idea of the price devel-
opment in the market (Grant, 1983). The Swiss Baumeister Verband (SBV)
went a bit further. It organized pre-tender meetings at which prices were
opened without names, in order to give participants an insight into the
price level. Furthermore, members negotiated over technical specica-
tions of the work and various calculation bases, and could agree on
binding tariffs for raw materials and wages (Kriesi and Schaller, 1984).
Germany has the strictest anti-cartel regime. Even the less far-reaching
452
regulations are illegal there. The Bundeskartellamt actively prosecutes

contractors who engage in collusive practices. In 1975, 336 German con-
tractors were ned 36 million marks and in 1983 seventy-seven rms
were ned 54 million marks (Czarnovski et al., 1976). It was the largest
economic scandal in the history of the Federal Republic. That alone is an
indication of how high the pressure in construction must be to mitigate
the erce competition. Informal and illegal price collusion probably takes
place everywhere in construction. The Dutch have preferred to take such
market agreements out of the secretive atmosphere of smoky backrooms,
to formalize them and thus to make them more subject to control.
The cartel may offset market failures such as ruinous competition and
imbalances of power. But might it not make matters worse by raising
prices for consumers and decreasing the quality of goods, as economists
would expect? To investigate this, we have compared the Dutch construc-
tion industry on a number of criteria with those of other EU and former
EFTA countries. The results are presented in Tables 7 and 8. The numbers
in parentheses are the rank order scores per variable.
Table 7 indicates that Dutch contractors do not make excessive prots
because of their cartel. The ‘rest income quota’ (non-labour income (rent,
interest and prots) as a percentage of value added) as an indicator of
protability in Dutch construction is the lowest among eight of the EU
countries. In Germany, with its strict anti-cartel policy, contractors make
almost twice as much prot.
The cartel does not discourage technical innovation, as Olson main-
tains for distributional coalitions. On the contrary, the Dutch construction
industry is among the most technologically advanced in Europe, judging
by its high labour productivity which is – like growth of labour produc-
tivity – among the highest in Europe.
Furthermore, the construction cartel does not make Dutch contractors
lazy and self-sufcient in the domestic market. Related to the size of the
sector, the Dutch are by far the most active abroad.
Neither does the construction cartel lead to exploitation of consumers.
Table 8 indicates that its building prices are among the lowest in Europe.
Only the less developed countries of Portugal and Greece are cheaper.
In the densely populated country Netherlands the average rent for a
comfortable three-room apartment in a large city is one of the lowest
in Europe. Except for Portugal and Greece, only the thinly populated
welfare state of Sweden is cheaper. The quality of housing in the
Netherlands is also high. Almost all houses have a bath or shower, which
is far from the case elsewhere in Europe, and the average housing age
in the Netherlands is one of the lowest in Europe.
In general, the Dutch construction industry is the top scorer on a number
of indicators of static and dynamic efciency. The cartel seems to increase
competitive power by mitigating ruinous competition. The cartel also
453
Table 7 Prots and productivity in the construction industry: the Netherlands

compared with other European countries
Country Rest income Labour Annual labour Construction abroad

as % of value productivity productivity change in US$ per domestic
added in ecus in construction in employee in
(1987) (1989) relation to average construction
for all industry (1989)
(1979–89)
Netherlands 4.1 (8) 29,800 (3) 1.1 (1.5) 5,924 (1)

EEC:
Belgium 4.8 (7) 34,600 (1) –1.5 (7) 3,711 (2)
Germany 7.2 (3) 25,360 (4) 0.0 (5.5) 569 (10)
Denmark n.a. 24,200 (7) 0.2 (3) 1,636 (7)
Spain 16.2 (1) 18,872 (10) 0.0 (5.5) 390 (11)
France 5.3 (5) 24,975 (5) 0.1 (4) 3,493 (3)
UK 6.8 (4) 21,100 (9) –2.4 (9) 2,030 (5)
Greece n.a. n.a. n.a. n.a.
Italy 11.6 (2) 6,420 (11) –3.7 (10) 1,434 (8)
Ireland 5.2 (6) n.a. n.a. n.a.
Portugal n.a. 4,080 (12) 1.1 (1.5) n.a.
Non-EEC:
Austria n.a. 24,737 (6) –1.8 (8) n.a.
Switzerland n.a. 30,560 (2) n.a. 2,569 (9)
Sweden n.a. n.a. n.a. 1,909 (4)
Sources: Knechtel (1990, Tables 13, 19, 20, 23); OECD, Labour Force Statistics (Paris); authors’
calculations.
Note: The numbers in parentheses indicate the rank-order scores. Column 1: Rest income
is the sum of interest paid, insurance premiums and prots, in rms with twenty or more
employees. Column 2: Gross value added per worker. Data for Sweden are for 1985.
n.a. = not available.
brings a minimal stability to the market and the sector structure, bene-
cial for planning long-term investments. Abuse of market power by the
cartel is not so likely or frequent. The typical sector structure and tender-
ing procedure in construction still guarantee sufcient competition.
The example of construction indicates that a ‘distributional coalition’
such as a cartel will have the most positive and least negative conse-
quences for economic efciency when a reasonable balance can be found
between the demands of stability and exibility. Some degree of compe-
tition should be maintained. That is likely in sectors exposed to inter-
national competition or in sectors with large numbers of rms.
The construction cartel is just one example of the many cartels that
until recently organized the Dutch economy. In 1992 the ofcial Dutch
cartel register contained 245 market-sharing agreements, 267 price and
454
Table 8 Prices and quality in the construction industry: the Netherlands

compared with European countries
Country Building prices Average monthly % of houses Average age

for housing rent in US$ with bath or of houses
per m2 in ecus (1989) shower in years
(1987) (1981–7) (1985)
Netherlands 471 (3) 434 (4) 97 (3) 34.9 (2)

EEC:
Belgium 976 (12) 602 (9) 76 (8) 48.1 (11)
Germany 555 (6) 532 (6) 99 (1) 38.7 (4)
Denmark 588 (8) 473 (5) 87 (4) 44.5 (8)
Spain 548 (5) 1,090 (13) n.a. 34.2 (1)
France 577 (7) 963 (12) 85 (6.5) 44.1 (7)
UK 539 (4) 1,245 (14) 98 (2) 44.0 (6)
Greece 275 (2) 423 (3) n.a. n.a.
Italy 635 (9) 612 (10) 86 (5) 38.5 (3)
Ireland n.a. 554 (8) 85 (6.5) 46.5 (10)
Portugal 191 (1) 361 (1) 58 (9) 41.4 (5)
Non-EEC:
Austria 756 (11) 534 (7) n.a. 46.0 (9)
Switzerland 713 (10) 698 (11) n.a. n.a.
Sweden n.a. 416 (2) n.a. n.a.
Sources: Czerny (1990: 69; Austrian Schillings converted to ecus using exchange rates in
Knechtel, 1990); World Economic Forum (1992) for column 2; Knechtel (1990: Tables 7, 9).
Note: n.a. = not available. Column 2 gives the rent for an unfurnished three-room at in
a large city.
tendering agreements, 202 distribution agreements and 45 collective

exclusive treading agreements, the latter mostly of professional organ-
izations which limited market entry (OECD, 1993: 60). Hence the OECD
has called Dutch competition law ‘the unusual case’. The Netherlands
was one of the few countries to have had an ‘abuse’ regime since the
1930s. That is, cartels were legal, unless their destructive consequences
could be demonstrated. Recently Dutch law has been changed to bring
it into line with the European Union and other member states where
the opposite principle holds. But until 1997 the Netherlands could be
considered a ‘cartel paradise’. This system was widely accepted, even
by customers, as proved by the scarcity of formal complaints.
The cartels do not seem to have had a detrimental effect on the econ-
omy. The country ranked high while the cartels were still in place, for
example, in the World Competitiveness Report of the International
Institute for Management Development and the World Economic Forum
in Lausanne (World Economic Forum, 1992). In 1991 it ranked seventh,
455
after other countries with ‘organized capitalism’ such as Japan, Germany,

Austria and Switzerland. In 1996 Holland still ranked seventh. Even such
a liberal forum as the OECD had to admit that ‘collusive agreements’
such as the Dutch cartels did not seem to harm the economy. Even though
many sectors of the economy are enmeshed in a web of restrictive
agreements, regulations, and barriers to entry . . . while economic
growth has been slightly lower than the EC average over the last
decade or so, prices are not obviously higher than in most other
countries. What is the answer to this apparent paradox? A possible
explanation might be that the Dutch way of dealing with compe-
tition, combined with the openness of the economy, has been
sufcient to prevent most blatant abuses of market power; also, the
social consensus may have had a benecial impact on overall
economic performance.
(OECD, 1993: 57)
In the sectors discussed associations played an important role in
reducing uncertainty and encouraging growth. Here the net effects were
clearly positive. Of course, this does not always have to be the case.
Such sectoral comparisons indicate that one cannot – at the very least –
generalize Olson’s arguments about the negative effects of associations.
6 S EC TO R A L C O R PO R A T ISM A N D
MA C R O EC O N O MI C G R O W TH :
A C OU N TR Y C O M PA R ISO N
Associational activity at the sectoral level may enhance the performance
of those sectors. But how would it affect the performance of economies
as a whole? What is good for one sector may not necessarily be good
for others or for the economy as a whole. Sectors could ‘externalize’ their
costs to other sectors.
In this respect the distinction between macro- and meso-corporatism
is relevant. Macro- or intersectoral corporatism refers to the organiza-
tion of economic interests in large encompassing (peak) associations and
hence to governance arrangements that involve intersectoral concerta-
tion between peak associations of labour, capital and sometimes the state,
usually on the issue of macroeconomic policy, including overall wage
development. Meso- or sectoral corporatism refers to well-organized
associations at the sectoral level that perform governance functions for
that sector. This can be in the eld of wage bargaining, but in order to
be considered real meso-corporatism associations need to perform gover-
nance functions in other elds as well, such as the ones addressed above:
product quality regulation, vocational training, collective research and
development, and general market ordering.
456
There is an extensive literature on the economic performance of macro-

corporatist arrangements (e.g. Schmitter, 1981; Czada, 1983; Cameron,
1984; Lehmbruch, 1984; Lehner, 1988; Calmfors and Drifll, 1988; Crepaz,
1992; Pekkarinen et al., 1992). These authors show that encompassing
associations have positive effects on economic performance indicators
such as low strike incidence, high productivity, low ination and gradual
adjustment to external shocks and technological change. The rationale
behind these ndings might be that, as Olson has argued, encompassing
interest associations have more problems in externalizing costs and
pursuing particularistic policies. That is, these ndings are still in line
with Olson’s thesis. But does Olson’s argument on the negative effects
of non-encompassing special-interest associations also hold true?
It is not easy to characterize countries according to the prevalence of
sectoral or meso-corporatism. We have done so in several ways. First,
we have based ourselves on the classications in the literature on the
dominant level of wage negotiations in several countries (most recently
Traxler, 1996; Visser, 1998; see also Table 6). Second, we have based
ourselves on the OBI data on the involvement of business associations
in public policy (which provided a qualitative but not a quantitative
indication) and on other case studies on the prevalence of sectoral corpo-
ratism in different countries. Third, we asked twelve established scholars
in the eld (Schmitter, Lehmbruch, Czada, Grant, Heritier, Schmid,
Kriesi, Crouch, Traxler, Visser, Hemerijck, Grande) to classify countries
according to the prevalence of meso-corporatism. On the basis of this
information we come to the following classication of countries on the
prevalence of meso-corporatism.
High sectoral corporatism, or intensive delegation of public policy to
sectoral associations and self-regulation by industry, is found in Austria,
Germany, Belgium, the Netherlands and Switzerland. Eight of the twelve
mentioned ranked these countries high on the meso-corporatism scale.
A number of comparative policy studies and country monographs,
among others produced in the OBI project, show that sectoral associa-
tions are very much involved in self-regulation and that the state has
delegated the formulation and/or implementation of public policy to a
variety of private interest associations. Such arrangements are found both
in a diversity of important sectors (small business, Handwerk, construc-
tion, agriculture, banking and insurance) and in various policy elds
(social security, health, competition policy, vocational training industrial
policy, research and development, environmental policy). It is not coin-
cidental that these countries are, perhaps with the exception of Austria,
located on the central city-belt of late medieval Europe (Rokkan, 1981),
where bourgeois merchant oligarchies developed an early tradition of
self-government by organizations of civil society which survived the
onslaught of the French Revolution, and which at present is manifested
457
in the chamber and guild organizations found in these countries. Thus

in these societies, forms of self-organization had emerged from civil soci-
ety at an early date.
Medium sectoral corporatism is found in the Scandinavian countries of
Sweden, Norway, Denmark and Finland. These countries have also a
tradition of self-organization by civil society, for example, in Sweden
based on the early presence of independent farmers. However, these
countries have a stronger etatist tradition, which also resulted in strong
centralization of interest associability. This was in part because of their
absolutist inheritance, but more so because of the dominance of social
democracy. Therefore there was less delegation of authority to private
associations and less tolerance for self-regulation, particularly of busi-
ness, because this was sometimes perceived as a threat to workers’ and
consumers’ interests. These countries score high on macro-corporatism,
which also reduces the leeway for sectoral associations. In addition,
Italy was classied by most experts as being medium on the scale of
meso-corporatism, which ts both with indications from case studies
and with its location on the European city-belt stemming from the
Renaissance.
Weak or no sectoral corporatism is found in countries with a liberal-
pluralist tradition (the UK, USA, Ireland), in countries where the political
culture as well as the public administration are highly suspicious of
‘special interest’ associations (USA), and in France with its strong etatist
tradition and dislike of particularistic interests, which have been
perceived as threats to the interests of the ‘great French state’ and/or
as political vehicles of the Catholic Church (Hayward, 1975). This suspi-
cion of interest associations dates back to the French Revolution – which
set out to destroy all intermediary organs between the citizen and the
state – and to the problematic history of Church–state relations in that
country (Crouch, 1993).
Institutions of sectoral corporatism have been typical of these coun-
tries for longer periods of time. Many have their roots in the process of
state formation there. Others have been created in response to the
economic crises of the 1870s and 1930s (Katzenstein, 1985). They have
also played an important role in the edice of the welfare state, erected
after 1945. Hence, this classication can be considered to hold for the
whole pre- and post-war period under study.
In Table 9 we classify fourteen countries according to the prevalence
of sectoral corporatism. The table compares their average national growth
rates over different periods of time. Over the period 1950 to 1994 coun-
tries that score high on sectoral corporatism have on average performed
best, although the differences are small. Their average annual growth
rate is 3.4 per cent, higher than the categories of medium- and low
sectoral corporatist countries with on average 3.1 per cent and 3 per cent.
458
Table 9 Average annual growth rates of countries classied by the prevalence

of sectoral corporatism
1929–38 1938–48 1948–58 1958–73 1973–81 1981–94 1950–94
High sectoral corporatism:

Austria –1.8 –0.9 7.6 4.7 2.5 2.2 3.6
Germany 3.5 –6.2 7.3 4.7 1.9 2.2 4.0
Netherlands –2.2 6.7 5.8 4.9 2.0 2.1 3.4
Belgium, nnp 3.7 0.3 3.2 4.7 2.0 2.0 3.1
Switzerland n.a. 2.5 3.5 4.6 0.5 1.7 2.9
Group average 3.4
Medium sectoral corporatism:
Denmark 2.2 1.7 3.1 4.5 1.2 2.2 2.8
Finland, ndp 3.7 0.5 4.2 5.3 2.5 2.0 3.6
Norway 2.6 n.a. 3.5 4.1 4.2 2.4 3.5
Sweden 2.2 3.3 3.4 4.1 1.5 1.2 2.6
Italy 1.6 –0.1 5.9 5.2 3.3 2.0 3.8
Group average 3.1
Low sectoral corporatism:
France, nnp –1.9 –0.4 4.3 5.2 2.4 2.2 3.6
Ireland, ndp 0.1 1.3 0.4 4.3 4.4 3.3 3.4
United Kingdom 1.8 1.5 2.4 3.3 0.6 2.1 2.2
USA n.a. n.a. n.a. 3.9 2.0 2.2 2.9
Group average 3
Total average 1.2 0.9 4.3 4.6 2.2 2.1 3.3

Notes: Data in the rst three columns refer to growth of Gross Domestic Product at constant
prices and are from Flora et al. (1987) and the authors’ own calculations. For Belgium and
France only Net National Product (nnp) and for Finland and Ireland only Net Domestic
Product (ndp) data were available. The last four columns are OECD data of GDP at 1980
prices in US$ and authors’ own calculations. Average annual growth rates have been calcu-
lated according to the formula: (ln (xt)–ln(xo))/t.
A cautious conclusion would be that countries scoring high on sectoral

corporatism do at least not perform worse with regard to growth.1 Even
if we give no econometric proof that associations improve growth rates,
our data indicate that associations are, at the very least, not harmful to
growth as Olson presumes them to be.
Some might argue that, in the long run, institutional arrangements are
not the decisive factor for growth. Technical variables relevant to the
production function, such as growth of labour inputs and investment
ratios, technical change or economies of scale, may be more important
for explaining long-term growth. Van de Klundert and van Schaik (1993)
found normal patterns of growth for eight countries between 1870 and
1989 primarily determined by investment ratios.
459
But what then determines high investment ratios? Eichengreen (1994)

showed, for Western Europe after the Second World War, that high
investment ratios were largely due to institutions singularly well suited
to reconstruction and growth. They solved commitment and coordination
problems. In particular, they disseminated information, monitored the
compliance of interest groups with their agreements to moderate wage
claims and boosted investment. Furthermore, Eichengreen showed that
some countries had a poorer growth performance because they failed to
develop the relevant domestic institutions or did so only with delay.
The countries he mentions are the United Kingdom, Ireland, France and
Italy. With the exception of Italy, these are precisely those European
countries that we rank as low sectoral corporatist countries in Table 9.
Furthermore, associational governance could also positively inuence
other economic performance indicators. Schettkat (1992) showed that the
same growth rates of GDP in Germany and the USA were accompanied
by different growth rates of labour productivity. The ‘quality’ of growth
was hence quite different between countries. The German institutional
setting achieved high labour productivity increases (through the voca-
tional training system, the varying of working hours instead of hiring
and ring, and through demand management), while American growth
was mainly due to an extension of the (partly low-qualied) service
sector. Even though in the long run both countries show very similar
GDP growth rates, associational institutional settings increased the
quality of growth.
C O NC L U S IO N
Mancur Olson’s books are appealing for their logical consistency and
parsimony. They provoke debate and invite confrontation with empir-
ical data. This article makes that confrontation and nds the data
contradict the three major hypotheses of the 1982 book.
1 Contrary to Olson’s assumptions, the number of associations does not
increase over time. Furthermore, contrary to what he suggests, war
does not break up ‘distributional coalitions’ but tends to strengthen
them.
2 The argument that the greater the number of special interest associ-
ations the worse does not hold. The greater the number of special-
interest associations, the more opposition will develop among them
and the lower the chances that public policy will be dominated by
them.
3 ‘Distributional coalitions’, or interest associations which pursue partic-
ularistic interests, do not necessarily threaten growth. They do not
always hamper the introduction of technological innovations or the
460
optimal allocation of input factors through the erection of market

entry barriers.
Even the purely redistributive activities of interest associations can
have positive economic effects. Union demands for wage increases in
line with productivity growth and for social security programmes have
increased or at least stabilized domestic demand over time, thus
providing Keynesian demand stimulation of the economy. Fordism
would perhaps not have been possible without unions – and without
employers’ associations as their partners in negotiations.
However, associations in many European countries are concerned not
only with redistribution but also with improving allocation. Even the
creation of market entry barriers can have favourable economic effects
by reducing ruinous competition, correcting imbalances of power on
markets, dampening uctuations in prices, and reducing excessive
volatility in turnover. Associational regulation may be important for
engendering trust and stabilizing expectations, thus enhancing long-term
investment. Furthermore, associations may provide public goods in the
form of research and development, vocational training, etc., which reduce
temporary market failures and uncertainties and thus allow for long-
term planning and growth of the economy.
The positive functions of associational regulation may be offset by the
‘regulatory failures’ which Olson had in mind. Associational lobbying
and regulation can of course contribute to structural rigidities and inef-
ciencies, such as oversized research and development units, an
improperly trained workforce, incentives to rms to avoid risk taking.
There is often a trade-off between the reduction of uncertainty and
the safeguarding of exibility (cf. also Traxler and Unger, 1994b).
Associations can provide stability, but this stability may become rigidity.
Whether the one or the other dominates depends partly on the charac-
teristics and problems of the sector. There are, however, indications that
overall the balance strikes in favour of associations. A comparison of
long-term growth data shows that countries with sectoral associational
governance have on average higher overall economic growth rates.
NO T E
1 We do not control for the inuence of variables other than the degree of
sectoral corporatism on growth rates. Other variables can be found in Denison
(1985) who analysed how much different sources of growth contributed to
growth rate differentials in nine western countries between 1850 and 1962.
He studied sources of growth in labour input (employment, hours of work,
age and sex composition of workers, education), in capital input (dwellings,
earnings on international assets, non-residential structure and equipment,
inventories), and growth in land. As indicators of technological progress he
distinguished advances in knowledge, changes in the lag in application of
461
knowledge, and improved allocation of resources. He found that growth rate

differences between countries were largely determined by the extent to which
inputs were increased, by the reallocation of resources from farm to non-
farm employment, and by economies of scale (1985: 326ff.).
R E F E R EN C E S
Aldrich, Howard and Staber, Udo (1988) ‘Organizing business interests: patterns
of trade association foundings, transformation, and death’, in Glenn Carroll
(ed.) Ecological Models of Organizations, Cambridge, Mass.: Ballinger, pp. 111–26.
Aldrich, Howard, Staber, Udo, Zimmer, Catherine and Beggs, John (1990)
‘Minimalism and organizational mortality: patterns of disbanding among US
trade associations, 1890–1983’, in Jitendra Singh (ed.) Organizational Evolution.
New Directions, Newbury Park, pp. 21–52.
Aldrich, Howard, Zimmer, Catherine, Staber, Udo and Beggs, John (1991) ‘Trade
association foundings in 20th century America: economic differentiation,
government actions, and population dynamics’, revised version of a paper
presented at the conference of the European Group of Organizational
Sociologists (EGOS) in Vienna.
Barclays Bank (1990) Construction Survey, London.
Bardach, Eugene and Kagan, Robert A. (1982) Going by the Book: the Problem of
Regulatory Unreasonableness, Philadelphia, Pa.: Temple.
Barnouw, David and Nekkers, Jan (1991) ‘State corporatism against the state’,
in Wyn Grant, Jan Nekkers and Frans van Waarden (eds) Organizing Business
for War, Oxford: Berg.
Bernstein, Marver (1955) Regulating Business by Independent Commission, Westport,
Conn.: Greenwood.
Buchanan, James M., Tollison, Robert D. and Tullock, Gordon (eds) (1980) Towards
a Theory of the Rent-Seeking Society, Texas.
Calmfors, Lars and Drifll, John (1988) ‘Bargaining structure, corporatism and
macroeconomic performance’, Economic Policy 6: 14–61.
Cameron, David (1984) ‘Social democracy, corporatism, labor quiescence, and
the representation of economic interest in advanced capitalist society’, in
John Goldthorpe (ed.) Order and Conict in Contemporary Capitalism, Oxford:
Oxford University Press, pp. 143–78.
—— (1988) ‘Distributional coalitions and other sources of economic stagnation:
on Olson’s Rise and Decline of Nations’, International Organization 42(4)
(Autumn).
CBS (Central Bureau of Statistics of the Netherlands) (1998) ‘Leden vakorgan-
isaties’, Sociaal-economische maandstatistiek 01: 52–62.
Chandler, Alfred D. (1980) ‘Government versus business: an American phenom-
enon’, in John D. Dunlop (ed.) Business and Public Policy, Cambridge, Mass.:
Harvard University Press.
Coleman, William D. (1988) Business and Politics. A Study of Collective Action,
Kingston.
—— (1990) ‘State traditions and comprehensive business associations: a compar-
ative structural analysis’, Political Studies 38: 231–52.
Crepaz, Markus M. (1992) ‘Corporatism in Decline? An empirical analysis of the
impact of corporatism on macroeconomic performance and industrial
Disputes in 18 industrialized democracies’, Comparative Political Studies 25:
139–68.
Crouch, Colin (1993) Industrial Relations and European State Traditions, Oxford.
462
—— (1995) ‘Organized interests as resources or as constraint: rival logics of voca-

tional training policy’, in Colin Crouch and Franz Traxler (eds) Organized
Industrial Relations in Europe: What Future?, Aldershot: Avebury, pp. 311–30.
Cuff, Robert (1973) The War Industries Board: Business–Government Relations during
World War I, Baltimore, Md.: Johns Hopkins University Press.
Czada, Roland (1983) ‘Konsensbedingungen und Auswirkungen neokorporatis-
tischer Politikentwicklung’, Journal für Sozialforschung 23(4): 421–39.
Czarnovski, Jürgen von, Gutzle, Helmut, Hoffmann, Volker and Strauch,
Volkmar (1976) ‘Kartelle in der Bauindustrie. Eine Studie über Erscheinungs-
bild, Aufklärung und Verminderung von Straftaten und Ordnungs-
widrigkeiten in der Bauwirtschaft’, Technologie und Politik Aktuell Magazin 5
(June): 57–140.
Czerny, Margarete (1990) ‘Internationaler Vergleich’, in Margarete Czerny (ed.)
Gutachten zur Neugestaltung der Wohnungspolitik in Oesterreich, Vienna: WIFO-
Austrian Institute of Economic Research.
Denison, Edward F. (1985) Why Growth Rates Differ, Postwar Experience in Nine
Western Countries, Washington, DC.
Eichengreen, Barry (1994) ‘Institutions and economic growth: Europe after World
War II’, CEPR Discussion Paper no. 973, London.
Flam, Helena (1990) ‘Corporate actors: denition, genesis, and interaction’,
Discussion Paper of the Max-Planck-Institut für Gesellschaftsforschung, Cologne
(November).
Flora, Peter, Kraus, Franz and Pfenning, Winfried (1987) State, Economy, and
Society in Western Europe 1815–1975. A Data Handbook, Vol. II: The Growth of
Industrial Societies and Capitalist Economies, Frankfurt.
Geer, Hans de (1991) ‘Corporatism and neutrality: Sweden during the Second
World War’, in Wyn Grant, Jan Nekkers and Frans van Waarden (eds)
Organizing Business for War, Oxford: Berg.
Grant, Wyn (1983) ‘The organization of business interests in the UK construc-
tion industry’, Discussion Paper IIM/LMP 83–25, Berlin: Wissenschafts-
zentrum Berlin.
—— (1987) Business and Politics in Britain, Houndsmill and London: Macmillan.
Grant, Wyn, Nekkers, Jan and van Waarden, Frans (eds) (1991) Organizing
Business for War. Industrial Policy and Corporatism during the Second World War,
Oxford: Berg.
Hannan, Michael. and Freeman, John H. (1988) ‘The ecology of organizational
mortality: American labor unions, 1836–1985’, American Journal of Sociology
94: 25–52.
Hayward, Jack (1975) ‘Employers’ associations and the state in France and
Britain’, in Steven Warnecke and Ezra Suleiman (eds) Industrial Policies in
Western Europe, New York.
Hollingsworth, Rogers, Schmitter, Philippe C. and Streeck, Wolfgang (eds) (1994)
Governing Capitalist Economies. Performance and Control of Economic Sectors,
Oxford.
Hoogenboom, Ari and Hoogenboom, Olive (1976) A History of the ICC. From
Panacea to Palliative, New York: Norton.
Jacobs, D., Boekholt, Peter and Zegveld, Wouter. (1990) De economische kracht van
Nederland. Een toepassing van Porter’s benadering van de concurrentiekracht van
landen, The Hague: TNO Beleidsstudies.
Kagan, Robert A. (1990) ‘How much does law matter? Labor law, competition,
and waterfront labor relations in Rotterdam and US ports’, Law and Society
Review 24(1): 35–69.
463
—— (1991) ‘Adversarial legalism and American government’, Journal of Policy

Analysis and Management 10: 369–406.
Kagan, Robert A. and Axelrad, Lee (1996) ‘Adversarial legalism: costs and conse-
quences’, unpublished draft, Center for Law and Society, University of
California, Berkeley.
Katzenstein, Peter J. (1985) Small States in World Markets. Industrial Policy in Europe,
Ithaca, NY: Cornell University Press.
Kelman, Steven (1981) Regulating America, Regulating Sweden: a Comparative Study
of Occupational Safety and Health Policy, Cambridge, Mass.
Klundert, Theo van de and Schaik, Anton van (1993) ‘On the historical conti-
nuity of the process of economic growth’, paper presented at the CEPR
workshop ‘Interpreting Postwar Growth’, Berlin, June.
Knechtel, Erhard F. (1990) Die Bauwirtschaft in der EG. Unternehmen im interna-
tionalen Vergleich. Daten – Fakten – Kommentare, Wiesbaden: Hauptverband
der Deutschen Bauindustrie.
Kriesi, Hanspeter and Schaller, Roland (1984) Wirtschaftsverbaende im Schweizer
Bauhauptgewerbe Zurich: Research Report, Soziological Institute University of
Zurich.
Lehmbruch, Gerhard (1984) ‘Concertation and the structure of corporatist
networks’, in John H. Goldthorpe (ed.) Order and Conict in Contemporary
Capitalism. Studies in the Political Economy of Western European Nations, Oxford.
—— (1986) ‘Die Messung des Einusses von “Verteilungskoalitionen ”: Zu
Weedes Ueberpruefung von Olsons Stagnationshypothese ’, Politische Viertel-
jahresschrift 27: 415–19.
Lehner, Franz (1988) ‘The political economy of distributive conict’, in Francis
Castles, Franz Lehner and Manfred Schmidt (eds) Managing Mixed Economies,
Berlin and New York.
Lindberg, Leon N., Campbell, John L. and Hollingsworth, J. Rogers (1991)
‘Economic governance and the analysis of structural change in the American
economy’, in Governance of the American Economy, Cambridge: Cambridge
University Press, pp. 1–34.
Lipset, Seymour Martin (1996) American Exceptionalism. A Double-Edged Sword,
New York and London: Norton.
Luyten, Dirk (1991) ‘The “policy of the lesser evil”. Corporatist economic orga-
nization in Belgium’, in Wyn Grant, Jan Nekkers and Frans van Waarden
(eds) Organizing Business for War, Oxford: Berg.
Maddison, Angus (1995) Explaining the Economic Performance of Nations. Essays in
Time and Space, Aldershot.
Maier, Charles S. (1981) ‘Fictitious bonds . . . of wealth and law’, in Susanne D.
Berger (ed.) Organising Interests in Western Europe, Cambridge, Mass.:
Cambridge University Press, pp. 26–61.
Marcus, Alfred A. (1984) The Adversary Economy. Business Responses to Changing
Government Requirements, Westport, Conn.
McConnell, Grant (1966) Private Power and American Democracy, New York.
Mitnick, Barry M. (1980) The Political Economy of Regulation, New York.
Moe, Terry M. (1980) The Organization of Interests. Incentives and the Internal
Dynamics of Political Interest Groups, Chicago.
OECD (1976) Collusive Tendering, Paris.
—— (1978) Comparative Summary of Legislations on Restrictive Business Practices,
Paris.
—— (1992) Historical Statistics 1960–1990, Paris.
—— (1993) Economic Surveys. The Netherlands, Paris.
464
Olson, Mancur (1965) The Logic of Collective Action. Public Goods and the Theory
of Groups, Cambridge, Mass.
—— (1982) The Rise and Decline of Nations. Economic Growth, Stagation and Social
Rigidities, New Haven, Conn.: Yale University Press.
—— (1983) ‘The political economy of comparative growth rates’, in Dennis C.
Mueller (ed.) The Political Economy of Growth, New Haven, Conn.:Yale
University Press.
Pekkarinen, Jukka, Pohjola, Matti and Rowthorn, Bob (eds) (1992) Social
Corporatism. A Superior Economic System?, Oxford.: Oxford University Press.
Rokkan, Stein (1981) ‘Territories, nations, parties: toward a geoeconomic-
geopolitical model for the explanation of variations within Western Europe’,
in R. L. Merritt and B. Russett (eds) From National Development to Global
Community, London, pp. 70–95.
Rossiter, Adrian (1991) ‘Business and corporatism in Vichy France’, in Wyn
Grant, Jan Nekkers and Frans van Waarden (eds) Organizing Business for
War, Oxford: Berg.
Sander, Richard (1992) ‘Elevating the debate on lawyers and economic growth’,
in Law and Social Inquiry 17.
Schettkat, Ronald (1992) The Labor Market Dynamics of Economic Restructuring, The
United States and Germany in Transition, New York.
Schmitter, Philippe C. (1981) ‘Interest intermediation and regime governability
in contemporary Western Europe and North America’, in Susanne D. Berger
(ed.) Organizing Interests in Western Europe. Pluralism, Corporatism and the
Transformation of Politics, Cambridge, pp. 285–327.
—— (1990) ‘Sectors in modern capitalism: modes of governance and variations
in performance’, in Renato Brunetta and Carlo Dell’Aringa (eds) Labor
Relations and Economic Performance, Houndmills.
Schmitter, Philippe C. and Streeck, Wolfgang (1981) ‘The organization of
business interests. A research design to study the associative action of
business in the advanced industrial societies of Western Europe. Revised
and extended version’, Discussion Paper IIM/LMP 81–31, Berlin: Wissen-
schaftszentrum Berlin.
Schubert, Klaus (ed.) (1992) Leistungen und Grenzen politisch-oekonomischer Theorie.
Eine kritische Bestandsaufnahme zu Mancur Olson, Darmstadt: Wiss. Buch-
gesellschaft.
Shapiro, Martin (1997) ‘The problems of independent agencies in the United
States and the European Union’, Journal of European Public Policy 4(2) (June):
276–91.
Stinchcombe, Arthur L. (1965) ‘Organizations and social structure’, in James G.
March (ed.) Handbook of Organizations, Chicago, pp. 142–93.
Streeck, Wolfgang (1989) ‘The territorial organization of interests and the
logics of associative action: the case of Handwerk organization in West
Germany’, in William D. Coleman and Henry J. Jacek (eds) Regionalism,
Business Interests and Public Policy, London and Newbury Park: Sage, pp.
59–94.
Traxler, Franz (1996) ‘Collective bargaining and industrial change: a case of disor-
ganization? A comparative analysis of eighteen OECD countries’, European
Sociological Review 12(3): 271–87.
Traxler, Franz and Unger, Brigitte (1994a) ‘Industry or infrastructure? A cross-
national comparison of governance, its determinants and economic
consequences in the dairy sector’, in Rogers Hollingsworth, Philippe
Schmitter and Wolfgang Streeck (eds) Governing Capitalist Economies,
Oxford.
465
—— (1994b) ‘Governance, economic restructuring and international competi-

tiveness. On the relationship between efciency and regulation from a
theoretical and empirical view’, Journal of Economic Issues (March).
Vinke, Harriet and Wilthagen Ton (1992) The Non-Mobilization of Law by Asbestos
Victims in the Netherlands: Social Insurance versus Tort-Based Compensation,
Amsterdam: Hugo Sinzheimer Institute.
Visser, Jelle (1989) European Trade Unions in Figures: 1913–1985, Deventer.
—— (1998) Social Dialogue and Industrial Relations in Austria, Denmark, Ireland and
the Netherlands, report prepared for the ILO Country Employment Policy
Review in Selected OECD Countries, Amsterdam (November).
Visser, Jelle and Hemerijck, Anton (1997) ‘A Dutch Miracle’. Job Growth, Welfare
Reform and Corporatism in the Netherlands, Amsterdam: Amsterdam University
Press.
Visser, Jelle and Waddington, Jeremy (1996) ‘Industrialization and politics: a
century of union structural development in three European countries’,
European Journal of Industrial Relations 2(1): 21–53.
Vogel, David (1978) ‘Why businessmen distrust the state: the political conscious-
ness of American corporate executives’, British Journal of Political Science 8:
45–78.
—— (1986) National Styles of Regulation. Environmental Policy in Great Britain and
the United States, Ithaca, NY: Cornell University Press.
—— (1989) Fluctuating Fortunes. The Political Power of Business in America, New
York: Basic Books.
Waarden, Frans van (1985a) ‘Varieties of collective self-regulation of business:
the example of the Dutch dairy industry’, in Wolfgang Streeck and Philippe
Schmitter (eds) Private Interest Government. Beyond Market and State, Beverly
Hills, Calif. London and New Delhi: Sage, pp. 197–220.
—— (1985b) ‘Regulering en belangenorganisaties van ondernemers’, in F. L. van
Holthoon (ed.) De Nederlandse samenleving sinds 1815, Assen and Maastricht:
Van Gorcum, pp. 227–60.
—— (1987a) Het geheim van Twente. Fabrikantenverenigingen in de oudste grootin-
dustrie van Nederland 1800 – 1940, Amersfoort and Leuven: Acco.
—— (1987b) ‘Problemen van machtsvorming in een heterogene werkgeversv-
ereniging. De beginjaren van de Algemene Werkgeversvereniging (AWV)
(1919–1925) als voorbeeld’, Tijdschrift voor Arbeidsvraagstukken 4: 16–27.
—— (1989) Organisatiemacht van belangenverenigingen. De ondernemersorganisaties
in de bouwnijverheid als voorbeeld, Amersfoort and Leuven: Acco.
—— (1991a) ‘Two logics of collective action? Business associations as distinct
from trade unions: the problem of associations of organizations’, in Otto
Jacobi and Dieter Sadowski (eds) Employers’ Associations in Europe, Baden
Baden: Nomos, pp. 51–84.
—— (1991b) ‘Wartime economic mobilization and state–business relations. A
comparison of eight countries’, in Wyn Grant, Jan Nekkers and Frans van
Waarden (eds) Organizing Business for War, Oxford: Berg.
—— (1992a) ‘Emergence and development of business interest associations. An
example from the Netherlands’, Organization Studies 13(4): 521–61.
—— (1992b) ‘Mancur Olsons Theorie und die Empirie des Kollektiven Handelns.’,
in Klaus Schubert (ed.) Leistungen und Grenzen politisch-ökonomischer Theorie:
Eine kritische Bestandsaufnahme zu M. Olson, Darmstadt: Wiss. Buchverlag,
pp. 139–68.
Weber, Hajo (1991) ‘Political design and systems of interest intermediation:
Germany between the 1930s and 1950s’, in Wyn Grant, Jan Nekkers and
Frans van Waarden (eds) Organizing Business for War: Oxford, Berg.
466
Weede, Erich (1986) ‘Verteilungskoalitionen, Staatstaetigkeit und Stagnation’,

Politische Vierteljahresschrift 27: 222–36.
Wilson, Graham (1986) The Politics of Safety and Health, Oxford.: Oxford University
Press
Windmuller, J. P. (1970) Arbeidsverhoudingen in Nederland, Utrecht and Antwerp:
Spectrum.
World Economic Forum (1992) The World Competitiveness Report 1991, Lausanne:
IMD.
467

Unger and Van Waarden 1999

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Unger and Van Waarden 1999

Uploaded by

Copyright:

Available Formats

Review of International Political Economy 6:4 Winter 1999: 425–467

Interest associations and economic

Review of International Political Economy

explain variations in economic growth rates by reference to institutions,

study on the organizational structure and activities of 352 business inter-

1 Interest associations are bad. They

reduce efciency and aggregate income in the societies in which

3.1 Time and numbers

strong uctuations; we will return to that – and a certain stabilization

Table 1 Foundings of nationally relevant voluntary trade associations in the

Period Netherlands Canada USA

Sweden Netherlands Britain

1892–1900 45 2 3 55.8 112 3 4 76.1 603 322 166 1196.3

Source: Visser and Waddington (1996).

members represents a more important resource, and workers usually

3.2 War and disbanding of associations

Table 3 Average age of business interest associations in nine countries

Number of associations Average age in

Relatively old associations:

Austria have on average even somewhat older associations. Many of

3.3 War and crises as creators rather than

representatives’, particularly when they lost their statutory tasks after

3.4 War and encompassing organizations

Table 4 Concentration of the union movement: number of peak union

Country Number of peak Union Total union member-

1933 1939 1945 1985 1933 1939 1945 1985

4.1 Numbers and political inuence

confronted with interests from many economic sectors. By balancing

4.2 Asymmetry in the political arena?

As by Olson’s assumption, more associations emerge over time, the

access, recognition, inuence, concessions, etc., thus uprooting possibly

do assist production and efcient allocation. They do so in particular

5.1 Information asymmetries, opportunism, distrust:

however, still in the hands of private interest associations. Similar devel-

5.2 Reduction of transaction costs of commercial

and lengthy court battles. These transaction costs may be considered

Table 5 Density of lawyers and tort cases by country

Country Lawyers Lawyers per Tort costs as %

United States 780,000 312.0 2.4

channelled into institutionalized consultation and collective bargaining,

5.3 Short-termism and the problem of providing collective goods

Table 6 Associational collective bargaining and strike frequency

Category and country % workers covered by Working days lost to strikes

I Northern corporatism: centralized bargaining:

A special case arises where it is difcult to exclude others from sharing

Germans are so famous. This system of trade associations is responsible

5.4 Ruinous competition: cartels in the construction industry

industry, where up to 40 per cent of rms disappear every year (Barclays

regulations are illegal there. The Bundeskartellamt actively prosecutes

Table 7 Prots and productivity in the construction industry: the Netherlands

Country Rest income Labour Annual labour Construction abroad

Netherlands 4.1 (8) 29,800 (3) 1.1 (1.5) 5,924 (1)

Table 8 Prices and quality in the construction industry: the Netherlands

Country Building prices Average monthly % of houses Average age

Netherlands 471 (3) 434 (4) 97 (3) 34.9 (2)

tendering agreements, 202 distribution agreements and 45 collective

after other countries with ‘organized capitalism’ such as Japan, Germany,

There is an extensive literature on the economic performance of macro-

in the chamber and guild organizations found in these countries. Thus

Table 9 Average annual growth rates of countries classied by the prevalence

1929–38 1938–48 1948–58 1958–73 1973–81 1981–94 1950–94

High sectoral corporatism:

Total average 1.2 0.9 4.3 4.6 2.2 2.1 3.3

A cautious conclusion would be that countries scoring high on sectoral

reduce efciency and aggregate income in the societies in which

strong uctuations; we will return to that – and a certain stabilization

4.1 Numbers and political inuence

access, recognition, inuence, concessions, etc., thus uprooting possibly

do assist production and efcient allocation. They do so in particular

A special case arises where it is difcult to exclude others from sharing

industry, where up to 40 per cent of rms disappear every year (Barclays

Table 7 Prots and productivity in the construction industry: the Netherlands

Table 9 Average annual growth rates of countries classied by the prevalence